Investigating the temporal dynamics of topic variation, consistency, and diversity in funding and funded papers: evidence from Chinese LIS projects

Research funding undergoes thorough peer review to ensure alignment with research goals before approval. However, significant discrepancies between funded research topics and resulting publications may indicate issues such as misattribution of funding or disruptive innovations. Traditional evaluation methods based on publication counts are insufficient for assessing topic consistency. This study presents a quantitative analysis of the temporal dynamics of topic variation, consistency, and diversity between scientific funding and publications of Information science & library science (LIS) in China. A Dirichlet-multinomial regression (DMR) topic model was used to analyze topic variation while word-embedding and keyword clustering techniques measured topic consistency and diversity across funder categories, author affiliation, and funding duration. The results showed that higher education institutions exhibited varying levels of thematic consistency and diversity in their funded papers. “Double First-Class” universities exhibited high diversity and good consistency, while vocational colleges demonstrated the highest consistency. Over time, all institutions improved their research consistency, reflecting better fund utilization. National-level funds maintained high consistency while promoting diversity, whereas ministry and local-level funds balanced depth and scope, and institutional funds focused on internal needs. The National Natural Science Foundation of China supported a broader range of topics than the National Social Science Fund of China and Humanities and Social Science Major Engineering Projects, which focused on specific research areas. Projects that exceeded their planned timelines tended to exhibit lower consistency, highlighting the need for better management and accurate registration of project details. These insights guide the optimization of scientific fund management and alignment between funding and research outcomes.